Academic R&D and S&E educational activities have long been a significant part of the U.S. R&D enterprise. R&D spending by universities and colleges is projected to reach $23.8 billion in 1997, accounting for an estimated 12 percent of total national R&D expenditures. The academic sector also continues to be the single largest performer of basic research, accounting for an estimated 52 percent of national basic research expenditures. The bulk of funding for academic R&D is provided by the Federal Government (60 percent in 1997); the second largest funding source is higher education institutions themselves (19 percent). State and local governments contribute 8 percent of the total, and industry and all other sources combined account for about 7 percent each. The bulk of federal funding is provided by three agencies: the National Institutes of Health with 57 percent, the National Science Foundation with 15 percent, and the Department of Defense with 10 percent.
Extensive physical infrastructure exists in support of academic R&D. About $3.1 billion in expenditures for constructing new research facilities were planned for 1996-97, along with another $1.3 billion for repair and renovation. Since 1988 (when comparable data first became available), academic S&E research space has increased by 22 percent, to 136 million net assignable square feet. New construction projects initiated between 1986 and 1995 which will either replace existing or add new space are expected to produce over 52 million square feet of research space by the time they are completed. In 1996, deferred construction or renovation projects totaled $9.3 billion, of which $7.4 billion was carried on approved construction plans. The major facilities funding sources are state governments (38 percent) and the institutions themselves (23 percent). Expenditures for research equipment were running just below 6 percent of total 1995 R&D expenditures. The major funder of research equipment remains the Federal Government (59 percent in 1995). In 1996, academic institutions rated 37 percent of their research laboratory space as suitable for the most scientifically competitive research; 44 percent as possibly needing some repair or renovation but effective for most levels of research; and 19 percent as needing major repair, renovation, or replacement. Overall, 27 percent of in-use research instruments were judged to be state of the art, another 63 percent as adequate for researcher needs, and 9 percent as inadequate.
About half of the nation's doctoral S&E research workforce was located in academic institutionsroughly 153,000 in 1995, including postdoctorates. The number of academic doctoral scientists and engineers reporting research as their primary work responsibility continued to grow, reaching 83,000 in 1995. Much of the growth, especially since the mid-1980s, occurred outside the traditional research universities; the number of institutions in this segment with federal R&D support reached 654 in 1995, up from 335 in 1975. In the course of their work, academic researchers are supported by, and help train, about 330,000 full-time S&E graduate students. For about 90,000 of them, a research assistantship was their most important means of support in 1995. The Federal Government is the primary source of support for about half of these students. In fact, RAs have grown in importance. The proportion of graduate students with research assistantships as their primary means of support increased from 22 to 27 percent between 1980 and 1995. A larger percentage of graduate students in the physical sciences, the environmental sciences, and engineering rely on RAs as their primary mechanism of support than do students in other disciplines.
Academic researchers produced 71 percent of all U.S.-authored scientific and technical articles in an international core set of peer-reviewed natural science and engineering journals included in the Institute for Scientific Information's Science Citation Index, and 23 percent of the world output published in these journals. (The total U.S. article share in 1995 was 33 percent.) Academic scientists and engineers increasingly collaborate with colleagues elsewhere: in 1995, nearly a quarter of all academic articles involved one or more authors from another U.S. employment sector.
Academic research, though predominantly basic, is increasingly connected with potential practical applications. More than 1,800 patents were awarded to academic institutions in 1995, which represented over 3 percent of all U.S. patent grants in that year. Academic patents were concentrated in a smaller set of application areas than patents of other awardees, with significant strengths in the life sciences, physics, and chemistry. In fact, more than a quarter of all academic patents fell into only three application areas with presumed biomedical relevance. Income from patenting and licensing agreements continued to grow and reached $299 million in 1995. And the number of citations to scientific and technical articles on patent applications, which has risen strongly in recent years, exceeded 47,000 in 1996roughly 26,000 of which were to academic articles.
The increasingly global nature of the scientific and engineering enterprise is reflected in an ubiquitous increase in the number of articles that have authors from more than one country. Roughly half of the 439,000 articles published worldwide in the SCI journal set referred to earlier had authors from multiple institutions, and nearly 30 percent of these multi-author papers involved international collaboration. Two complementary trends characterize the U.S. position. For almost every nation with strong international coauthorship ties, the number of papers involving U.S. researchers rose strongly over the past decade and a half. But during this period, many nations broadened the reach of their international collaborations, leading to a gradual diminution of the U.S. share of articles involving international collaborations.
Citations to scientific and technical articles offer an indication of the perceived utility of the results of previous work in subsequent research. In a given country's literature, citations to local work tend to figure prominently and have less of a
time lag than citations to work published abroad. But U.S. authors tend to be cited by scientists in virtually all mature scientific nations in excess of the U.S. world share of articles in chemistry, physics, biomedical research, and clinical
medicine; U.S. articles in the remaining fields tend to be cited at or slightly below the U.S. share. But no other country cites the domestic literature as heavily as the United States67 percent in 1995probably reflecting, at least in
part, the sheer scale of the nation's scientific and technical enterprise.